Generally, a semiconductor manufacturing apparatus comprises: a load port into which a transfer container containing a plurality of substrates is loaded; a substrate transfer mechanism configured to draw the substrates from the transfer container and to return the processed substrates to the transfer container; and a processing part configured to perform various processes to the substrates. As the transfer container, there is used, instead of a conventional open-type carrier, a FOUP (Front Opening Unified Pod) having a lid member for opening and closing a front surface thereof. Thus, the semiconductor manufacturing apparatus is provided with a mechanism for opening and closing the lid member of the FOUR
As a process for a substrate, there are a single-wafer process, such as a vacuum process, and a coating and developing process (a coating process of a resist and a developing process after exposure), and a batch process, such as a thermal process by a vertical thermal processing apparatus and a substrate cleaning process. When a single-wafer process is performed, a substrate is generally drawn from a transfer container placed in a load port. On the other hand, when a batch process is performed, a transfer container is temporarily stored in a container storage part, which is called “stocker” provided between a load port and a processing part, so that stagnation of the transfer containers on the load port can be prevented, whereby a process can be effectively performed.
In a cleaning apparatus of a batch type, for example, a batch process is performed such that fifty semiconductor wafers (hereinafter referred to as “wafers”) are arranged in a cleaning container, and the wafers are sequentially immersed into a plurality of cleaning tanks. By improving a mechanism that draws the substrates from the transfer container and transports the substrates to the cleaning container, a series of processes have been recently accelerated. Although there is developed an apparatus whose throughput (process capability) is about six hundred substrates per hour, a still higher throughput is required. For example, assuming that the transfer container can contain twenty-five substrates, in order to process nine hundred substrates per hour, thirty-six transfer containers should be loaded into a load port from a transfer apparatus, specifically, e.g., an overhead transfer apparatus (an in-plant transfer apparatus of a overhead traveling type: OHT), which is disposed on a factory. In this case, the number of times of the loading operations and the unloading operations is seventy-two (36×2).
The number of stages for the transfer containers aligned in the load port is generally three or four. However, in order to cope with the above requirement, the number of stages aligned in the load port should be increased to, e.g., about eight. However, when the number of stages aligned therein is increased, a width of the apparatus is widened. Since a rear side area of the lord port is a dead space, the dead space is added to an installation area. As a result, because of the increased installation space, such a structure cannot be employed.
JP2008-277764A describes the following technique. Namely, a waiting position, in which a plurality of FOUPs can wait is provided on an upper surface of a ceiling part of a substrate processing apparatus. A FOUP supplied from an overhead transfer apparatus is temporarily located in the waiting position, and then the FOUP is moved by a moving mechanism to a support plate member above a load port. Thereafter, the FOUP is loaded into the load port from the support plate member by the overhead transfer apparatus.
JP2008-263004A describes the following structure. Namely, in a CVD apparatus of a single-wafer type, there are formed three load ports by arranging three lower FOUP stages, which can be moved upward and downward, on a front side of three pod openers (lid-member opening and closing mechanism). In addition, three upper FOUP stages are disposed on an upper surface of a sealing of the apparatus body. Thus, FOUPs can be transported between the upper FOUP stages and the lower FOUP stages.
Although these techniques can prevent stagnation of the FOUPs in the load port, these techniques have limits to achieving a still higher throughput as described above. Thus, a technique for more improving a throughput has been desired.